Sex-associated patterns of disease activity and physical function in ankylosing spondylitis: real-world cross-sectional evidence

Objective To characterize sex-associated differences in disease activity and physical function in ankylosing spondylitis (AS) and to formally test whether the cross-sectional BASFI–ASDAS association differed by sex. Methods We conducted a real-world cross-sectional analysis of 530 Chinese outpatients with rheumatologist-documented AS (387 men and 143 women). Between-sex distributions of BASDAI, BASFI, ASDAS-CRP, and ASDAS-ESR were compared using two-sided Mann–Whitney U tests, Cliff’s delta, and Holm adjustment. Sex-specific Spearman correlations, bootstrap differences in rho, and sex×ASDAS interaction models assessed whether the BASFI–ASDAS association differed by sex. Analyses were repeated in a 567-patient sensitivity cohort. Results In the primary cohort, ASDAS-CRP was modestly higher in men than in women (men-minus-women Cliff’s delta, 0.142; 95% CI, 0.033 to 0.248; raw P = 0.0119; Holm-adjusted P = 0.0476). No multiplicity-supported between-sex differences were identified for BASDAI, BASFI, or ASDAS-ESR (Holm-adjusted P = 0.257, 0.257, and 0.713, respectively). The men-minus-women differences in the BASFI–ASDAS Spearman correlations were 0.088 (95% CI, −0.108 to 0.289) for ASDAS-CRP and 0.087 (95% CI, −0.107 to 0.293) for ASDAS-ESR. Neither the correlation-difference tests nor the sex×ASDAS interaction models provided evidence that the BASFI–ASDAS association differed by sex, and results were concordant in the n=567 sensitivity cohort. Conclusion Sex-associated variation in this Chinese outpatient cohort was modest and outcome-specific. ASDAS-CRP was modestly higher in men, whereas BASDAI, BASFI, and ASDAS-ESR showed substantial overlap. Formal analyses did not identify sex modification of the BASFI–ASDAS association, supporting joint interpretation of disease activity and physical function as complementary clinical dimensions.